Myself and my lifelong friend (and newly joined maker) @Nicola_Papadia are dusting off an old hobby: building and flying model rockets.
After a 13-year break, this should be perfectly straightforward and absolutely nothing could possibly go wrong… right?
This time we’ve decided to make things unnecessarily complicated by adding an electronic payload.
The current plan includes a battery, Arduino, barometric sensor, SD card for data logging, possibly a gyroscope module, and ideally some sort of camera.
Unfortunately we’ve hit a bit of a wall with the Arduino programming side of things, so we’re looking for a friendly coding wizard to point us in the right direction. Pretty please!
Off the top of my head I’m thinking of @Kyle, @scday94, @mia, @davidN but I’ve also seen plenty of suspiciously talented coders gathered around the clean area tables.
If anyone fancies trading a bit of programming help for a pint, eternal gratitude, and the chance to contribute to a questionable aerospace programme, we’d love to hear from you.
I’ll leave it to Nico to explain exactly which modules we’re looking at, as he’s the electronics expert in this operation.
Launch is currently planned for the first half of August, so there’s absolutely no pressure whatsoever…
Also, as @Nicola_Papadia lives in Italy, he is not a SLMS member YET…
So please don’t move this conversation in the private area or he will loose access
Hi everyone! Essentially, we are trying to put together an arduino nano, a barometric sensor ( the most used seems to be BMP280), an sdcard module in order to save flight data. I tried and tried to write a sketch but it seem that this problem goes beyond my skills. Please help us!
I’m a software guy, I’ve done some Arduino programming for keyboards before… Rockets should be similar, right? Probably simpler, they have less buttons. I’ve logged over 100 hours in Kerbal Space Program anyway so we should be good.
What do you need programming? I’d love to help, always wanted to build a rocket. But I’m a bit hesitant before I know the rough scope.
Presumably the big requirements are that the system is low-latency (and robust)? Does it need to do anything fancy as far as gimbaling, or making any decisions as to where the thrust is being vectored?
Don’t be mate; even in case of complete electronic failure, the main mission objective is the noise and the flames! A simple lift-off will make us happy, so there is no pressure at all
If you use an esp I would hope you will be getting real time flight data and video over wifi.
BTW I used to build rockets years ago so will be very interested.
+1 for using an ESP32 instead of an Arduino (you could still use the Arduino IDE to program it). ESP32-CAM is cheap, small. RandomNerd tutorials has examples form many sensors
If you all needs some help I worked in the space industry for a while.
More then the programming your fist problem will be acceleration and the forces involved, you might want to calculate the possibile acceleration that the electronics will be subject at and check every component to figure out if they will be functioning at that speed (some MEMS sensore Will literally die if you overstress them)
My choice would be the Pico2 , mostly because you cal leverage the programmable units inside to do complex operations fast and also for the included boost converter that let is operate down to 1.2v.
Hi! I love the idea. I’m not a rocket scientist, but…
How are you going to use the collected data? If for machine learning, the more datapoints you can collect during the flight, the more accurate attributions you’ll be able to make to the events occuring. Sampling quality and rates will be direcltly corelated to the throughput bandwith limitations of the microcontrollers you will use. Memory writing speed is likely to be a bottleneck on do-it-all esp32 +if you decide you want more sensors (i.e. for fuel mass displacement, airflow, temperatures across components, accelerometry)= corrupted or low quality data. If I was planning it, I would be looking to design of the flight data analysis model first, evaluate the key computable unknowns for the projects next iteration, calculate the required bandwidth and select a microcontroller (or a set) with identified data bottlenecking factors to accomodate the throughput demand with a safe margin. I guess radio + memory hybrid would be the best approach, regardless as a fallback or complimentary logging. But I might be wrong
I can help with precision welding it you decide to use metals in the design.
